RNPC1, an RNA-binding protein and a target of the p53 family, is …genesdev.cshlp.org › content › 20 › 21 › 2961.full.pdf · 2006-11-01 · bility of p21 transcript is found

RNPC1, an RNA-binding proteinand a target of the p53 family, is requiredfor maintaining the stability of the basaland stress-induced p21 transcriptLimin Shu, Wensheng Yan, and Xinbin Chen1

Department of Cell Biology, The University of Alabama at Birmingham, Birmingham, Alabama 35294, USA

p21, a cyclin-dependent kinase inhibitor, is transcriptionally regulated by the p53 family to induce cell cyclearrest. p21 is also regulated post-transcriptionally upon DNA damage in a p53-dependent manner, but themechanism is uncertain. Here, we found that RNPC1, an RNA-binding protein and a target of the p53 family,is required for maintaining the stability of the basal and stress-induced p21 transcript. Specifically, we showedthat RNPC1 is induced by the p53 family and DNA damage in a p53-dependent manner. The RNPC1 geneencodes at least two alternative spliced isoforms, RNPC1a and RNPC1b, both of which contain an intact RNArecognition motif. Interestingly, we found that RNPC1a, but not RNPC1b, induces cell cycle arrest in G1,although both isoforms are expressed in the nucleus and cytoplasm. In addition, we found that while bothisoforms directly bind to the 3� untranslated region in p21 transcript, only RNPC1a is able to stabilize boththe basal and stress-induced p21 transcripts. Conversely, RNPC1a knockdown destabilizes p21 transcript.Finally, we found that RNPC1a is required to maintain the stability of p21 transcript induced by p53.

[Keywords: RNPC1; p21; p53; p53 family; transcriptional regulation; cell cycle]

Supplemental material is available at http://www.genesdev.org.

Received June 26, 2006; revised version accepted September 12, 2006.

Mutations of p53 tumor suppressor occur in >50% ofhuman cancer, and loss of p53 function is known to playa central role in cancer development (Vogelstein et al.2000). The importance of p53 in this process is under-scored by the ability of p53 as a transcription factor (Koand Prives 1996). Upon activation, p53 induces a numberof genes (Harms et al. 2004), such as p21(WAF1/CIP1)(el-Deiry et al. 1993), MDM2 (Barak et al. 1993; Wu et al.1993), IGFBP3 (Buckbinder et al. 1995; Harms and Chen2005), PIDD (Lin et al. 2000), and Killer/DR5 (Wu et al.1997), which mediate the diverse biological functions ofp53, including cell cycle arrest and apoptosis (Vousdenand Lu 2002).

RNA-binding proteins have been shown to post-tran-scriptionally regulate gene expression, such as polyade-nylation, RNA splicing, transport, stability, and transla-tion, all of which are emerging as critical mechanismsfor gene regulation in mammalian cells (Krecic andSwanson 1999). In addition, an increasing number ofRNA-binding proteins have been identified and found tobe altered in a variety of genetic disorders, such as fragile

X mental retardation and myotonic dystrophy (Siomi etal. 1993; Timchenko et al. 1996). In general, RNA-bind-ing proteins contains one or more RNA-binding motifs,such as RNA recognition motif (RRM), hnRNP K homol-ogy motif, RGG box, and dsRBD motif (Dreyfuss et al.1993, 2002; Krecic and Swanson 1999). Among theseRNA-binding motifs, RRM is the most prevalent type ofeukaryotic RNA-binding motifs (Dreyfuss et al. 1993).RRM is composed of two submotifs, RNP1 and RNP2(Dreyfuss et al. 1993). Among the RRM-containing pro-teins are the Elav-like RNA-binding protein family(Myer et al. 1997), which consist of HuB, HuC, HuD, andHuR (Gorospe 2003).

p21 is shown to be the major mediator of p53 to regu-late the cell cycle transition from G1 to S (el-Deiry et al.1993; Harper et al. 1993; Xiong et al. 1993). p21 is alsoregulated by other p53 family proteins; that is, p63 andp73 (Kaghad et al. 1997; Yang et al. 1998; Zhu et al.1998). In addition to the transcriptional regulation by thep53 family, the level of p21 is found to be regulated bypost-transcriptional mechanisms. For example, the sta-bility of p21 transcript is found to be increased in cellsirradiated with ultraviolet C (UVC) in a p53-dependentmanner (Gorospe et al. 1998) and in cells treated withvarious agents, such as tumor necrosis factor � (Shioharaet al. 1996), epidermal growth factor (EGF) (Giles et al.

1Corresponding author.E-MAIL [email protected]; FAX (205) 934-0950.Article published online ahead of print. Article and publication date areonline at http://www.genesdev.org/cgi/doi/10.1101/gad.1463306.

GENES & DEVELOPMENT 20:2961–2972 © 2006 by Cold Spring Harbor Laboratory Press ISSN 0890-9369/06; www.genesdev.org 2961

Cold Spring Harbor Laboratory Press on July 26, 2020 - Published by genesdev.cshlp.orgDownloaded from

http://genesdev.cshlp.org/

http://www.cshlpress.com

2003), prostaglandin A2 (Gorospe and Holbrook 1996),and phorbol myristate acetate (Liu et al. 1996). Uponexposure to UVC or treatment with EGF and prostaglan-din A2, HuR is translocated from nucleus to cytosol,which binds to the 3� untranslated region (UTR) in p21transcript and enhances its stability (Wang et al. 2000;Giles et al. 2003; Yang et al. 2004). Interestingly, thesubcellular shuttling of HuR protein following UVC ir-radiation is p53 independent (Wang et al. 2000). Simi-larly, HuD, a neuronal member of the Elav-like family, isnecessary for nerve growth factor-mediated neural differ-entiation, binds to and stabilizes p21 transcript (Josephet al. 1998). Recently, Msi-1, a member of the musashiRNA-binding protein family, is found to repress p21translation via directly binding to p21 transcript (Battelliet al. 2006).

Here we presented evidence that the RNPC1 gene,which encodes at least two isoforms of an RNA-bindingprotein through alternate splicing at the C terminus, isinduced by the p53 family and by DNA damage in ap53-dependent manner. While both isoforms contain oneRRM motif and bind to the 3� UTR in p21 transcript, thelarge isoform, RNPC1a, is capable of inducing G1 arrestand regulating the stability of p21 transcript. Further-more, we provided evidence that RNPC1a is required forthe maintenance of the stability of p21 transcript in-duced by p53. Together, we concluded that RNPC1 is atarget of the p53 family and serves as a mediator of thep53 family to regulate p21 post-transcriptionally.

Results

RNPC1 is a direct target of the p53 family

p21 is known to be regulated by the p53 family, andserves as a major mediator of the family to induce cellcycle arrest in G1. However, it is possible that othergenes that are regulated by the p53 family participate insuch a process. To do this, we performed a microarraystudy to identify potential common target genes for thep53 family. Among the genes up-regulated by variousp53 family proteins in MCF7 cells, RNPC1 appears to bea common target of the p53 family (Chen et al. 2003). Tofurther address this, Northern blot analysis was per-formed. We showed that RNPC1 was induced by wild-type p53 but not mutant p53(R249S) in both MCF7 andH1299 cells (Fig. 1A). The induction of p21 was mea-sured as a positive control and the level of glyceralde-hyde-3-phosphate dehydrogenase (GAPDH) measured asa loading control. Similarly, RNPC1 was induced byp63�, �Np63�, and p73� in both MCF7 and H1299 cells(Supplementary Fig. 1A).

DNA damage stabilizes and activates p53, leading toinduction of p53 target genes. We reasoned that RNPC1,as a p53 target gene, is likely to be induced by DNAdamage in cells that carry an endogenous wild-type p53gene. To this end, we used RKO, HCT116, and MCF7cell lines in which endogenous p53 is wild-type. We alsoused RKO-p53-KD and HCT116(p53−/−), in which endog-

enous wild-type p53 was knocked down and knockedout, respectively (Bunz et al. 1998; Liu and Chen 2006).We found that RNPC1 was induced in RKO, HCT116,and MCF7 cells treated with DNA-damaging agent doxo-rubicin or camptothecin (Fig. 1B, RNPC1 panel). Simi-larly, p21 was induced (Fig. 1B, p21 panel). However,little if any induction of RNPC1 and p21 was detected inRKO-p53-KD and HCT116(p53−/−) cells treated withdoxorubicin or camptothecin (Fig. 1B).

Next, we determined whether an increase in RNPC1transcript correlates with an increase in RNPC1 protein.RNPC1 is expressed as at least two alternative splicedisoforms, RNPC1a and RNPC1b. Polyclonal antibodyagainst histidine-tagged RNPC1 fusion proteins wasmade in rabbits and found to recognize HA-taggedRNPC1a, but not HA-tagged RNPC1b (SupplementaryFig. 1B). We found that the level of RNPC1a was signifi-

Figure 1. RNPC1 is induced by p53. (A) Northern blots wereprepared with RNAs purified from MCF7 and H1299 cells thatwere uninduced (−) or induced (+) to express p53 or p53(R249S).The blots were probed with cDNAs derived form the RNPC1,p21, and GAPDH genes, respectively. (B) Northern blots wereprepared with RNAs purified from RKO, RKO-p53-KD,HCT116, HCT116(p53−/−), and MCF7 cells that were mock-treated (−) or treated (+) with 0.4 µg/mL doxorubincin (DOX) or300 nM camptothecin (CPT) for 20 h. The blots were analyzedas in A. (C,D) Western blots were prepared using extracts fromMCF7 cells that were uninduced (−) or induced (+) to expressHA-tagged wild-type p53 or p53(R249S) or extracts from RKO orRKO-p53-KD cells that were untreated (−) or treated (+) with 0.4µg/mL doxorubincin (DOX). HA-tagged wild-type p53 andp53(R249S) in MCF7 cells were detected by anti-HA and endog-enous p53 in RKO cells detected by anti-p53 monoclonal anti-bodies. p21, RNPC1, and actin were detected by their respectiveantibodies.

Shu et al.

2962 GENES & DEVELOPMENT




cantly increased in MCF7 cells by wild-type, but notmutant, p53(R249S) (Fig. 1C, cf. lanes 2,4 and 1,3, respec-tively) and in RKO cells treated with doxorubicin in ap53-dependent manner (Fig. 1D, cf. lanes 2,4 and 1,3,respectively). Levels of p21 were measured as a positivecontrol, whereas levels of actin measured as a loadingcontrol. Similarly, we found that the levels of RNPC1awere increased in MCF7 cells by p63�, �Np63�, p63�,p73�, and �Np73�, but not �Np63� (Supplementary Fig.1C). To make certain that RNPC1b is induced by the p53family, RT–PCR was performed and showed that the lev-els of RNPC1b transcript were increased by p53, p63�,p73�, and �Np73� (Supplementary Fig. 1D).

p53 transactivates its target gene through binding tospecific DNA sequences in the promoter or intron. Thus,we searched for p53-responsive elements (p53REs) in thegenomic locus encoding the RNPC1 gene and found twopotential binding sites: one located at nucleotides −3351to −3330 in the promoter with the sequence ofAGCCTGGCTCCCAGCCTAGTCT, disignated asp53RE-1; and the other located at nucleotides +519 to+546 in intron 1 with the sequence of GGGCAAGTCCAGGCGGCCCCCCAAGCTC, disignated as p53RE-2.Upon alignment with the consensus p53RE (el-Deiry etal. 1992), we found three mismatches each in bothp53RE-1 and p53RE-2 (Fig. 2A, top panel; lowercase let-ters represent mismatches).

To examine p53RE-1, we cloned a promoter fragmentcontaining p53RE-1 from nucleotides −3424 to +85 and ashorter fragment lacking p53RE-1 from nucleotides−1215 to +85 into pGL2 luciferase reporter, respectively(Fig. 2A, bottom panel at the left). We found that p53 wasable to increase the luciferase activity under the controlof the RNPC1 promoter fragment from nucleotides−3424 to +85, but not under the control of the promoterfragment lacking the potential p53RE-1 from nucleotides−1215 to +85 (Fig. 2B). In contrast, mutant p53(R249S)was inert (Fig. 2B). Similarly, the RNPC1 promoter con-taining p53RE-1 was responsive to p63�, �Np63�, p63�,�Np63�, p73�, and �Np73� (Supplementary Fig. 2A).

To examine p53RE-2, a 379-base-pair (bp) fragment(nucleotides +373 to +751) from intron 1 (Fig. 2A, bottompanel at the right) was cloned upstream of the minimalc-fos promoter and a luciferase reporter in O-Fluc (Johan-sen and Prywes 1994). In addition, a deletion mutant(�520–546) that deletes the entire p53RE-2 and a pointmutant (C522A) that alters one of the critical nucleo-tides in p53RE-2 were also cloned into O-Fluc (Fig. 2A).We showed that the luciferase activity under the controlof the fragment +373/+751, but not the deletion mutant(�520–546) or point mutant (C522A), was substantiallyincreased by wild-type p53, p63�, p63�, and p73�, andlittle, if any, by mutant p53(R249S) (Fig. 2C; Supplemen-tary Fig. 2B, respectively).

To further analyze p53RE-1 and p53RE-2 in theRNPC1 gene, we performed chromatin immunoprecipi-tation (ChIP) assay with primers shown in Figure 2D.The binding of the p53 family proteins to p53RE-1 in thep21 promoter was also determined as a positive control(Fig. 2D). To analyze the binding of p53, RKO cells were

untreated (−) or treated (+) with doxorubicin to activatep53 followed by cross-linking with formaldehyde andimmunoprecipitation with anti-p53 antibody or anti-HAantibody as a control. We found that p53 bound to bothp53RE-1 and p53RE-2 in the RNPC1 gene as well as top53RE-1 in the p21 gene (Fig. 2E). To analyze the bindingof p63 or p73, MCF7 cells were uninduced (−) or induced(+) to express Myc-tagged p63� or HA-tagged p73� andthen used for ChIP assay. The p63–DNA complexes wereimmunoprecipitated with anti-Myc antibody along withanti-HA as a control. The p73–DNA complexes were im-munoprecipitated with anti-HA antibody along withanti-Myc as a control. We found that both p63� and p73�bound to p53RE-1 and p53RE-2 in the RNPC1 gene aswell as to p53RE-1 in the p21 gene (Supplementary Fig.

Figure 2. Identification of potential p53REs in the RNPC1gene. (A) Schematic presentation of the RNPC1 locus and lucif-erase reporter constructs along with two potential p53REs lo-cated in the promoter and intron I (p53RE-1 and p53RE-2). (B,C)The p53REs in the RNPC1 gene are responsive to p53 but notmutant p53(R249S). The luciferase assay was performed as de-scribed in Materials and Methods. (D) Schematic presentation ofthe RNPC1 promoter, intron 1, and the p21 promoter with thelocation of the p53REs and PCR primers used for ChIP assay. (E)p53 binds directly to the p53REs in the RNPC1 and p21 genes invivo. ChIP assay was performed as described in Materials andMethods.

p21 mRNA stability is regulated by RNPC1

GENES & DEVELOPMENT 2963




2C). In sum, these data suggest that RNPC1 is likely tobe a direct target of the p53 family.

RNPC1 is an RNA-binding protein

According to the published sequence (GenBank#BC018711), the RNPC1 gene encodes two RNPC1 iso-forms, designated RNPC1a with 239 amino acids andRNPC1b with 121 amino acids (Fig. 3A; SupplementaryFig. 3A). The sequence in RNPC1b is identical to theN-terminal region in RNPC1a. Both isoforms contain acanonical RRM (amino acids 35–107) (SupplementaryFig. 3A). Sequence alignment of RRMs among RNPC1and other RRM-containing proteins indicated that thetwo submotifs, RNP-1 and RNP-2, are highly conserved,matching all the consensus residues (Supplementary Fig.3A). The phylogenic analysis showed that RNPC1 be-longs to a family of RNA-binding proteins that consistsof musashi, HuR, U2AF65, and nucleolin (Supplemen-tary Fig. 3B).

Intracellular localization of RNPC1

As an RNA-binding protein, RNPC1 may be shuttling inand out of the nucleus along with its cargos, RNA tran-scripts. To test this, immunofluorescence microscopywas performed with an anti-RNPC1 antibody andshowed that endogenous RNPC1 was primarily localizedin the cytosol, but a minute amount of it was also de-tected in the nucleus (Supplementary Fig. 4A, top panel).To determine the cellular localization of RNPC1a andRNPC1b, both isoforms were HA-tagged at the N termi-

nus, which was then inducibly expressed in RKO cellsunder the control of a tetracycline-inducible promoter.Two representative cell lines each that can induciblyexpress HA-tagged RNPC1a or RNPC1b were shown inFigure 3B. Immunofluorescence microscopy showed thatlike endogenous RNPC1, RNPC1a and RNPC1b werelocalized primarily in the perinuclear membrance and/orcytosol (Supplementary Fig. 4A, middle and bottom pan-els). To further confirm this, whole-cell extracts and ex-tracts from the cytosol and nucleus were prepared inRKO cells inducibly expressing HA-RNPC1a or HA-RNPC1b. Consistent with the immunofluorescence as-say, both RNPC1a and RNPC1b were expressed in boththe cytosol and nucleus (Supplementary Fig. 4B). How-ever, the fraction of RNPC1b in the cytosol was morethan that in the nucleus (Supplementary Fig. 4B, cf. lanes8 and 12).

RNPC1a is capable of inducing cell cycle arrestwhen overexpressed

To determine the biological activity of RNPC1, RKO celllines that inducibly express either RNPC1a or RNPC1b(Fig. 3B) were used for growth rate analysis. We foundthat cell proliferation was suppressed by RNPC1a, butnot RNPC1b, over a 6-d testing period (Fig. 3C,E). Next,to determine whether the growth suppression is due tocell cycle arrest, apoptosis, or both, DNA histogramanalysis was performed. We showed that cells were ar-rested primarily in G1 by RNPC1a (Fig. 3D), but not byRNPC1b (Fig. 3F).

To rule out the possibility that the effect of RNPC1a iscell type specific, we generated MCF7, HCT116,

Figure 3. Overexpression of RNPC1a, but notRNPC1b, inhibits cell proliferation and inducescell cycle arrest in G1. (A) Schematic presenta-tions of the RNPC1 locus and the usage of exonsfor RNPC1a and RNPC1b. (B) Generation ofRKO cell lines that inducibly express HA-taggedRNPC1a or RNPC1b. The levels of HA-RNPC1aand RNPC1b were quantified with anti-HA. (C,E)The growth rate of RKO cells that were unin-duced or induced to express HA-tagged RNPC1a(C) or RNPC1b (E) was measured over a 6-d pe-riod. (D,F) RKO cells were uninduced (−) or in-duced (+) to express HA-RNPC1a or HA-RNPC1bfor 2 or 4 d and then stained with propidium io-dide for DNA histogram analysis as described inMaterials and Methods.

Shu et al.





HCT116(p53−/−), and HCT116(p21−/−) cell lines that in-ducibly express HA-tagged RNPC1a (Fig. 4A,D,F,H). Weshowed that cell proliferation was inhibited by RNPC1ain MCF7 and HCT116 cell lines (Fig. 4B,E). In addition,we showed that RNPC1a was able to suppress cell pro-liferation in p53- and p21-independent manners, sincep53 and p21 were somatically knocked out inHCT116(p53−/−) and HCT116(p21−/−), respectively (Fig.4G,I). Furthermore, DNA histogram analysis showed

that MCF7 cells were arrested in G1 upon induction ofRNPC1a (Fig. 4C).

RNPC1 directly binds to p21 mRNA and enhancesits stability

As an RNA-binding protein, RNPC1 is likely to exert itsactivity by regulating the function of a protein directlyinvolved in the cell cycle regulation. Since p21 is themost well-defined mediator of the p53 family in G1 ar-rest, we explored the possibility that the function ofRNPC1a is linked to that of p21. Thus, we performedWestern blot analysis and showed that the level of p21protein was significantly increased in RKO cells uponinduction of RNPC1a in a time-dependent manner (Fig.5A, p21 panel, cf. lanes 1,3,5,7 and 2,4,6,8, respectively).Similarly, the level of p21 was markedly increased inMCF7, HCT116, and HCT116(p53−/−) cells upon induc-tion of RNPC1a (Fig. 5B,C, p21 panels). However,RNPC1b had no effect on the level of p21 (Fig. 5D, p21panel, cf. lanes 1,3,5,7 and 2,4,6,8, respectively). Thesedata suggest that the effect of RNPC1a on p21 is not celltype specific. We also examined several other proteins,such as p53, PCNA, and p27 and found that none of theseproteins were significantly affected (Fig. 5A–C).

To determine whether RNPC1a is able to further in-crease the extent of p53 induction of p21, the level of p21was examined in RKO cells uninduced (−) or induced (+)to express RNPC1a for 3 d along with treatment of 0.4µg/mL doxorubicin (Fig. 5E) or 5 µM Nutlin-3 (Fig. 5F) for0, 4, 8, 12, or 24 h. Nutlin is an inhibitor of Mdm2, whichactivates p53 without causing DNA damage (Vassilev etal. 2004). We showed that RNPC1a was expressed uponinduction (Fig. 5E,F, RNPC1 panel), whereas p53 wasaccumulated upon treatment with 0.4 µg/mL of doxoru-bicin within 8 h (Fig. 5E, p53 panel) or 5 µM Nutlinwithin 4 h (Fig. 5F, p53 panel). Interestingly, the levels ofp21, but not mdm2, were further increased by RNPC1a,although both p21 and mdm2 were induced upon p53activation (Fig. 5E,F, p21 and mdm2 panels).

As an RNA-binding protein, RNPC1 may directly bindto p21 mRNA and then increase its stability. Thus,Northern blot analysis was performed and showed thatthe levels of p21 were increased in RKO cells upon in-duction of RNPC1a in a time-dependent manner (Fig.6A, p21 panel). To further test this, an RNA immuno-precipitation assay was performed using extracts fromRKO cells that transiently express HA-RNPC1a, HA-RNPC1a (F37S), or HA-RNPC1b. RNPC1a (F37S) carriesa serine instead of phenylalanine at codon 37, which islocated in the RNP2 submotif of the RRM and predictedto be critical for RNA binding (Supplementary Fig. 3A).Western blot analysis showed that all the proteins wereexpressed in RKO cells (Fig. 6B). Next, an anti-HA anti-body was used to immunoprecipitate potential RNPC1–RNA complexes, whereas an anti-Flag antibody was usedas control. The precipitated RNPC1–RNA complexeswere treated with RNase-free DNase to remove trappedgenomic DNAs, and then the pull-down RNAs were pu-rified by Trizol reagent for cDNA synthesis. Five per-

Figure 4. RNPC1a-mediated growth suppression is not celltype specific, and can be independent of p53 and p21. (A,D,F,H)Generation of MCF7, HCT116, HCT116(p53−/−), andHCT116(p21−/−) cell lines that inducibly express HA-RNPC1a.The level of HA-RNPC1a was quantified with anti-HA.(B,E,G,I) The growth rate of MCF7, HCT116, HCT116(p53−/−),and HCT116(p21−/−) cells that were uninduced or induced toexpress HA-RNPC1a was measured over a 6-d period. (C) MCF7cells were uninduced (−) or induced (+) to express HA-RNPC1afor 2 or 4 d and then stained with propidium iodide for DNAhistogram analysis.






cent of cell extracts were used directly for RNA extrac-tion as an input control. Upon RT–PCR amplification,p21 transcripts were found to associate with RNPC1aand RNPC1b, but not RNPC1a (F37S) (Fig. 6C, HA-IPpanel). No p21 transcripts were found to be immunopre-cipitated by control anti-Flag antibody (Fig. 6C, Flag-IPpanel). In addition, GAPDH transcripts were not foundto interact with RNPC1 (Fig. 6C, GAPDH column).

Next, we wanted to identify the region in p21 tran-script that is responsive to RNPC1a. To test this, weperformed a luciferase assay using pGL3 reporter thatcontains various regions from the 3� UTR in the p21transcript (Fig. 6D, left column; Giles et al. 2003). Wefound that upon cotransfection with RNPC1a, a signifi-cant increase of luciferase activity was observed forpGL3-1/7, pGL3-2/7, and pGL3-6/7, but not for the con-trol pGL3, pGL3-1/6 and pGL3-1215 (Fig. 6D, right col-umn). These data suggest that p21 mRNA stability islikely controlled via a potential cis-element locatedwithin nucleotides 879–1512 in the 3� UTR in p21 tran-

script, but not the previously identified AU-rich regionwithin nucleotides 571–829 bound by HuR and HuD (Jo-seph et al. 1998; Wang et al. 2000; Giles et al. 2003; Yanget al. 2004) or GTAGT within nucleotides 1819–1823bound by Msi (Battelli et al. 2006).

To further explore how RNPC1a increases the level ofp21 mRNA, Northern blot analysis was performed tomeasure the decay rate of p21 transcript in the presenceor absence of RNPC1a. Upon inhibition of new RNAsynthesis with actinomycin D, the relative half-life forp21 mRNA was more than doubled by RNPC1a from∼2.5 to 5.3 h (Fig. 6E,F).

RNPC1 is required for maintaining the stabilityof the basal and stress-induced p21 mRNA

To demonstrate the physiological relevance for the regu-lation of p21 by RNPC1a, we generated MCF7 cell linesin which endogenous RNPC1a can be inducibly knockeddown. Two representative cell lines were shown in Fig-

Figure 5. RNPC1a increases the level of p21expression in multiple cell types, which canbe independent of p53. (A–C) The level of p21protein is increased by RNPC1a. Westernblots were prepared using cell extracts fromRKO (A), MCF7 (B), HCT116 (C, lanes 1–8),and HCT116(p53−/−) (C, lanes 9–16) that wereuninduced (−) or induced (+) to express HA-RNPC1a for 12–72 h. HA-RNPC1a, p21, p53,p27, PCNA, and actin were quantified withanti-HA, anti-p21, anti-p53, anti-p27, anti-PCNA, and anti-actin, respectively. (D)RNPC1b has no effect on the level of p21 pro-tein. The experiment was performed as inA–C except that RKO cells were uninduced orinduced to express HA-RNPC1b. (E,F) Thelevels of p21 protein induced by p53 are fur-ther increased by RNPC1a. Western blotswere prepared with extracts from RKO cellsuninduced (−) or induced (+) to expressRNPC1 for 3 d along with treatment of 0.4µg/mL doxorubicin (E) or 5 µM Nutlin (F) for0, 4, 8, 12, or 24 h. HA-RNPC1a, p21, p53,mdm2, and actin were quantified with anti-HA, anti-p21, anti-p53, anti-mdm2, and anti-actin, respectively.

Shu et al.





ure 7A. Upon induction of small interfering RNA(siRNA) against RNPC1a for 2 or 3 d, the levels ofRNPC1a transcript were markedly reduced and the lev-els of p21 transcript were concomitantly decreased (Fig.7A, RNPC1a and p21 panels, cf. lanes 1,3,5,7 and 2,4,6,8,respectively). Next, we examined the decay rate of p21transcript and found that the relative half-life for p21mRNA was decreaed from ∼2.7 to ∼2.2 h (Fig. 7B,C). Wealso examined the levels of p21 protein upon knockdownof RNPC1a for 24–72 h. We found that the levels of p21protein, but not p53 and Mdm2, were decreased (Fig. 7D,cf. lanes 1,3,5 and 2,4,6, respectively). These data indi-cated that the stability of p21 transcript at a normal con-dition is controlled by RNPC1a.

Since p53 induction of p21 is enhanced by overexpres-sion of RNPC1a (Fig. 4E,F), we wanted to determinewhether p53 induction of p21 is attenuated by RNPC1aknockdown. To test this, two representative MCF7clones (#36 and #62) were treated with doxorubicin, Nut-lin, and camptothecin for 0–12 h in the presence or ab-sence of RNPC1a knockdown. We found that p53 accu-mulation was detected upon treatment with doxorubi-cin, Nutlin, and camptothecin, regardless of the status of

RNPC1a knockdown (Fig. 7E–G; Supplementary Fig.5A,B, p53 panels). However, upon treatment with Nutlinand camptothecin, p53 accumulation in one MCF7 clone(#36) was slightly attenuated in the absence of RNPC1awithin the first 3 h, but not at the later time points (Fig.7F,G, p53 panels, cf. lanes 3 and 4). Nevertheless, whilep21 was increased following p53 accumulation, the ex-tent of the p21 increase was substantially less in thepresence of RNPC1a knockdown than in the absence ofRNPC1a knockdown during the testing period (Fig. 7E–G; Supplementary Fig. 5A,B, p21 panels, cf. lanes 6,8,10and 5,7,9, respectively). Interestingly, p53 induction ofMdm2 and PolH, both of which are regulated by p53(Barak et al. 1993; Wu et al. 1993; Liu and Chen 2006),was not substantially affected by RNPC1a knockdown(Fig. 7E–G; Supplementary Fig. 5A,B, Mdm2 and PolHpanels). Taken together, these data suggest that RNPC1ais required for maintaining the stability of p21 inducedby p53.

Discussion

Many p53 family target genes have been identified, butto our knowledge, RNPC1 is the only common p53 fam-

Figure 6. RNPC1 directly binds to the 3� UTRin p21 transcript and enhances p21 mRNA sta-bility. (A) RNPC1a increases the level of p21mRNA. Northern blots were prepared usingRNAs purified from RKO cells that were unin-duced (−) or induced (+) to express HA-RNPC1afor 6–72 h. The blots were probed with cDNAsderived from p21 and GAPDH, respectively. (B)Expression of HA-RNPC1a, HA-RNPC1a (F37S),and HA-RNPC1b. A Western blot was preparedusing extracts from RKO cells that were tran-siently transfected with a pcDNA3 vector ex-pressing HA-RNPC1a, HA-RNPC1a (F37S), orHA-RNPC1b. The blot was probed with anti-HA. (C) HA-RNPC1a and HA-RNPC1b, but notHA-RNPC1a (F37S), bind directly to p21, but notGAPDH, transcripts. p21 and GAPDH tran-scripts were detected by RT–PCR from potentialRNPC1–p21 mRNA complexes immunoprecipi-tated with anti-HA, or a control complex immu-noprecipitated with anti-Flag as described inMaterials and Methods. (D) The 3� UTR betweennucleotide 879 and 1512 in the p21 transcript isresponsive to RNPC1a. H1299 cells were co-transfected with 200 ng of pGL3 control reporteror pGL3 reporter carrying various regions fromthe 3� UTR in p21 transcript along with emptypcDNA3 or pcDNA3 expressing HA-RNPC1a.Luciferase assay was carried out as described inMaterials and Methods. (E) The half-life of p21transcript is enhanced by RNPC1a. The levels ofp21 and GAPDH transcripts were measured byNorthern blot analysis using RNAs purifiedfrom RKO cells that were uninduced (−) or in-duced (+) to express HA-RNPC1a for 20 h, fol-lowed by treatment with 7.5 µg/mL of actino-

mycin D (Act D) for 0, 2, 4, 6, or 8 h. (F) The relative half-life of p21 transcript in the absence or presence of RNPC1a. Uponnormalization with the level of GAPDH transcript, the relative half-life of p21 transcript was calculated from three separate experi-ments.






ily target gene that encodes an RNA-binding protein.Here we showed that all the p53 family members candirectly transactivate RNPC1. Interestingly, when over-expressed, RNPC1a is capable of inducing cell cycle ar-rest in G1 at least in part via binding to and stabilizingp21 transcript. Conversely, RNPC1a knockdown de-creases the stability of p21 transcript. More importantly,we provided evidence that RNPC1a is required for themaintenance of the stability of p21 transcript induced byp53, and overexpression of RNPC1a further increases theextent of p53 induction of p21. While the stability of p21transcript can be regulated by HuR following UVC irra-diation, the activity of HuR is controlled by its subcel-lular shuttling from nucleus to cytosol, which is p53independent (Wang et al. 2000). Thus, HuR is unlikely tomediate p53-dependent stabilization of p21 transcript.As a p53 target gene, we hypothesize that RNPC1 isresponsible for p53-dependent stabilization of p21 tran-script, and therefore serves as both an effecter and a me-diator of the p53 family in inducing cell cycle arrest inG1 (Fig. 8).

The 3� UTR in p21 transcript has been shown to con-tain several recognition sequences for a group of RNA-

binding proteins. For example, HuR and HuD were foundto recognize the AU-rich elements located withinnucleotides 571–829 (Joseph et al. 1998; Wang et al.2000; Giles et al. 2003; Yang et al. 2004) and Msi wasfound to recognize GTAGT within nucleotides 1819–1823 (Battelli et al. 2006). CP1 was also reported to bindto the 3� UTR (Giles et al. 2003), but its binding site hasnot been mapped. Here, we found that the binding regionfor RNPC1 is located between nucleotide 879 and 1512,which contains several C-rich elements but lies outsidethe AU-rich (HuR and HuD) elements and the Msi-bind-ing site. However, the precise nucleotide sequence re-mains to be determined. Since HuR and AUF1 bind tomany common AU-rich target mRNAs, including p21,and exert opposing influence on target mRNA stability(Lal et al. 2004), it is possible that RNPC1 may competeand/or cooperate with other RNA-binding proteins, in-cluding HuR and Msi, to bind to their respective recog-nition sequences, and consequently, control the out-come for p21 mRNA stability.

We showed that both RNPC1a and RNPC1b can bindto the p21 mRNA. This is not surprising since both iso-forms contain an intact RNA-binding domain, RRM. In-

Figure 7. RNPC1 is required for maintainingthe stability of the basal and stress-inducedp21 mRNA. (A) Generation of two MCF7 celllines in which endogenous RNPC1a is induc-ibly knocked down using a tetracycline-in-ducible H1 promoter. Northern blots wereprepared using RNAs purified from MCF7cells that were uninduced (−) or induced (+) toexpress siRNA against RNPC1a for 2 d and3 d, respectively. The blots were probed withcDNAs derived from RNPC1, p21, andGAPDH, respectively. (B,C) The half-life ofp21 transcript is decreased by RNPC1aknockdown. The levels of p21 and GAPDHtranscripts were measured by Northern blotanalysis using RNAs purified from MCF7cells that were uninduced (−) or induced (+) toknock down RNPC1a for 2 d, followed bytreatment with 7.5 µg/mL of actinomycin D(Act D) for 0, 1, 2, 4, or 6 h. (D) RNPC1aknockdown decreases the basal level of p21protein in MCF7 cells. Western blots wereprepared using extracts from MCF7 cells thatwere uninduced (−) or induced (+) to knockdown RNPC1a for 24, 48, or 72 h. The blotswere probed with antibodies against p21,Mdm2, p53, and actin, respectively. (E–G)RNPC1a is necessary for the maintenance ofthe levels of p21 protein induced by p53 inMCF7-RNPC1a-KD-36. Western blots wereprepared using extracts from MCF7 cells thatwere uninduced (−) or induced (+) to knockdown RNPC1a for 48 h, followed by treat-ment with 0.4 µg/mL of doxorubincin (DOX)(E), 5 µM Nutlin (F), or 300 nM camptothecin(CPT) (G) for 0–12 h. The blots were probedwith antibodies against p53, p21, Mdm2,PolH, and actin, respectively.

Shu et al.





terestingly, we also showed that RNPC1a, but notRNPC1b, is capable of inducing cell cycle arrest and in-creasing the stability of p21 transcript. One possible ex-planation is due to the intracellular localization of thesetwo isoforms: While both isoforms are expressed both inthe nucleus and cytosol, the level of RNPC1b in thecytosol appears to be higher than that in the nucleus.Upon binding to p21 or other not yet identified tran-scripts, RNPC1a might facilitate their transport fromnucleus to cytosol wherein protein is translated, and atthe same time, further protect these transcripts fromdegradation by cytosolic RNases. Thus, the kinetics ofp21 mRNA stability can be controlled by the stoichiom-etry of RNPC1a versus RNPC1b. Since both isoformscontain an identical amino acid sequence in the N ter-minus, the unique 108 residues in RNPC1a may interactwith other proteins that together regulate p21 mRNAstability. However, RNPC1b may help RNPC1a to regu-late the stability of p21 or other not yet identified tran-scripts. Therefore, further studies are warranted to ad-dress these potential functions of RNPC1.

Due to the nature of binding to RNA transcripts,RNA-binding proteins often display multiple functions.For example, HuR is known to regulate several early re-sponse genes, which then regulate the cell cycle and cel-lular differentiation (Lopez de Silanes et al. 2004). Wealso showed that in addition to inhibiting cell prolifera-tion by stabilizing the p21 transcript, RNPC1a is able toinhibit cell proliferation in HCT116 cell line in whichp21 is somatically knocked out. This suggests thatRNPC1a regulates other RNA transcripts involved in thecontrol of cell proliferation. Thus, other RNPC1 targetsneed to be identified in future studies, which will pro-vide an insight into the mechanism by which RNPC1regulates other cell cycle control pathways and other un-known functions.

Materials and methods

Plasmids

HA-tagged RNPC1a and RNPC1b were amplified by PCR usingEST clone (GenBank #BC018711) as a template with the follow-

ing primers: sense primer #1 for both isoforms, 5�-GAAGCTTGCCGCCATGGAGTACCCATACGACGTACCAGATTACGCTATGCTGCTGCAGCCCGCGCCG-3�; antisense primer #2for RNPC1a, 5�-GGAATTCTCACTGCATCCTGTCAGGCTGC-3�; and antisense primer #3 for RNPC1b, 5�-GGAATTCTCAGCCCGTCTGGAGGCTCCGCG-3�. To generate RNPC1a(F37S), two-step PCR reactions were performed. The first-stepPCR reaction was performed to separately amplify two DNAfragments: Fragment #1, encoding the N-terminal region ofRNPC1a, was amplified with the above sense primer #1 andantisense primer #4, 5�-TACGGCAGGCCGCCCACGGAGATCTTGGTGAACGT-3�; and fragment #2, encoding the C-terminal region of the RNPC1a, was amplified with senseprimer #5, 5�-ACGTTCACCAAGATCTCCGTGGGCGGCCTGCCGTA-3�, and the above antisense primer #2. The second-step PCR reaction was performed using a mixture of fragments#1 and #2 as a template with sense primer #1 and antisenseprimer #2. These cDNAs were subcloned into pGEM T easyvector and confirmed by sequencing. A HindIII–EcoRI fragmentcontaining the entire coding region was then purified andcloned into pcDNA3 and pcDNA4 for transient and stable in-ducible expressions, respectively.

To generate a construct that expresses a siRNA againstRNPC1a under the control of the tetracycline-regulated H1 pro-moter, two 68-base oligonucleotides targeting RNPC1a weredesigned similarly as that for targeting p53 except that the tar-geting sequence is derived from RNPC1 transcript from nucleo-tides +568 to +588 (NM_017495) (Brummelkamp et al. 2002; Liuand Chen 2006). The pair of oligos was then annealed andcloned into pBabe-H1 at HindIII and BglII sites, and the resultingconstruct was designated pBabe-H1-siRNPC1a. pBabe-H1 is aPolIII promoter-driven vector with a tetracycline operon se-quence inserted before the transcription start site (van de We-tering et al. 2003).

To generate a luciferase reporter under the control of a poten-tial p53RE in the RNPC1 gene, the RNPC1 promoter fragmentsfrom nucleotides −3424 to +85 (the first nucleotide upstream ofthe putative transcription start site is designated as −1) and fromnucleotides −1205 to +85 were amplified by PCR and clonedseparately into pGL2. The fragments were confirmed by se-quencing. To generate a luciferase reporter under the control ofa potential p53RE in intron 1 of the RNPC1 gene, a fragmentfrom nucleotides +373 to +751 was amplified and cloned intothe O-Fluc reporter vector upstream of a c-fos basic promoterand a luciferase reporter (Johansen and Prywes 1994). Similarly,a deletion mutant (�520–546) that lacks the entire potentialp53RE and a point mutant (C522A) that carries a mutation atthe critical nucleotide necessary for p53 binding were generatedby PCR and then separately cloned into O-Fluc. Deletion andmutation were confirmed by sequencing. pGL3 vectors thatcarry various regions from the 3� UTR in p21 transcript werekindly provided by Drs. S. Lee and P. Leedman (Harvard MedicalSchool, Boston, MA, and Western Australian Institute for Medi-cal Research, Perth, Western Australia, Australia).

Cell culture

RKO, MCF7, and HCT116, were cultured in DMEM mediumsupplemented with fetal bovine serum. HCT116(p53−/−) andHCT116(p21−/−) are derivatives of HCT116 in which the p53and p21 genes were somatically knocked out, respectively(Waldman et al. 1996; Bunz et al. 1998); RKO-p53-KD is a de-rivative of RKO in which p53 was stably knocked down by RNAinterference (RNAi) (Liu and Chen 2006). The MCF7 cell lines,which inducibly express p53, p53(R249S), p63�, �Np63�, andp73�, and the H1299 cell lines, which inducibly express p53,

Figure 8. A model for the role of RNPC1 in the p53 familypathway to induce G1 arrest.






�Np63�, and p73�, were as previously described (Zhu et al.2000; Dohn et al. 2001; Nozell et al. 2003). RKO, MCF7,HCT116, HCT116(p53−/−), and HCT116(p21−/−) cell lines,which inducibly express HA-tagged RNPC1a or RNPC1b, weregenerated as previously described (Dohn et al. 2003). To gener-ate inducible RNPC1a knockdown cell lines, pBabe-H1-siRNPC1a was transfected into MCF7 cells in which a tetracy-cline repressor is expressed by pcDNA6 (Yan and Chen 2006).The RNPC1a knockdown cell lines were selected with puromy-cin and confirmed by Northern blot analysis.

Luciferase assay

A dual luciferase assay was performed in triplicate according tothe manufacturer’s instructions (Promega). Briefly, p53-nullH1299 cells were plated at 5 × 104 cells per well in a 24-wellplate and allowed to recover overnight. Cells were then cotrans-fected with 200 ng each of pGL2, O-Fluc, or pGL3 together withpcDNA3 or a pcDNA3 vector expressing wild-type p53, mutantp53(R249S), p63, p73, or RNPC1a. As an internal control, 5 ng ofpRL-CMV, a Renilla luciferase vector (Promega), was also co-transfected per well. Thirty-six hours post-transfection, lucifer-ase activity was measured with the dual luciferase kit andTurner Designs luminometer. The fold increase in relative lu-ciferase activity is a product of the luciferase activity induced bya p53 family protein or RNPC1a divided by that induced byempty pcDNA3 vector.

ChIP assay

ChIP assay was performed as previously described (Liu et al.2003; Harms and Chen 2005). RKO cells were mock-treated ortreated with doxorubicin for 24 h, whereas MCF7 cells wereuninduced or induced to express Myc-tagged p63� or HA-taggedp73� for 24 h. Chromatins in these cells were then cross-linkedwith 1% formaldehyde for 10 min at room temperature, soni-cated to generate 500- to 1000-bp DNA fragments, and immu-noprecipitated with anti-p53 antibody to capture p53–DNAcomplexes, with anti-myc antibody to capture myc-taggedp63�–DNA complexes, or with anti-HA antibody to captureHA-tagged p73�–DNA complexes. After reverse cross-linkingand phenol-chloroform extraction, the bound DNA fragmentswere purified by a Qiagen column. PCR was performed to visu-alize the enriched DNA fragments. The primers designed toamplify the p53RE within the RNPC1 promoter were sense,5�-CAGCACACACAGTTGCCAGTCC-3�, and antisense, 5�-CTTAGGTATCAGCTGACTGCCAAGG-3�. The primers de-signed to amplify the p53RE within intron 1 of the RNPC1 genewere sense, 5�-TTCCAAGCATAGCGTCGCAAACTC-3�, andantisense, 5�-CTCCTCCCGTCAAAGTTCAAGGC-3�. Theprimers designed to amplify the upstream p53RE within the p21promoter were as previously described (Liu et al. 2003).

DNA histogram analysis

Cells induced or uninduced to express or knock down RNPC1for various times were collected and fixed in 75% ethanol for atleast 1 h at 4°C. Cells were then washed with PBS and resus-pended in staining buffer with 100 µg/mL RNase A and 50 µg/mL propidium iodine (Invitrogen). The percentage of cells ineach phase of the cell cycle (G1, S, and G2–M) was analyzedby a FACS-Caliber cell sorter (BD Biosciences) along withCellQuest software.

Northern blot analysis

Total RNAs were isolated from H1299, RKO, MCF7, andHCT116 cells using Trizol reagent (Invitrogen). Northern blot

analysis and preparation of p21 and GAPDH probes were asdescribed previously (Chen et al. 1995). The RNPC1 probe, a2.1-kb Xho1–EcoR1 fragment, was made from an EST clone(BC018711).

Antibody production and immunoblotting

To prepare anti-RNPC1 antibody, the full-length RNPC1a wascloned into a pRSet-A bacterial expression vector (Invitrogen).His-tagged RNPC1 protein was expressed in bacteria and puri-fied with Ni-agarose beads. Anti-RNPC1 antibody was raised intwo rabbits. For immunoblotting, cells were collected from cul-ture plates in PBS, resuspended in 2× sample buffer, and boiledfor 7 min at 95°C. Anti-RNPC1 antibody was diluted in PBSTcontaining 5% BSA instead of 5% dry milk. Other antibodiesused were anti-p53 monoclonal antibodies (DO-1, PAb1801,PAb240, and PAb421); anti-actin (Sigma); anti-p21 (C-19) (SantaCruz Biotechnology); anti-PCNA (Santa Cruz Biotechnology);anti-Mdm2 (Santa Cruz Biotechnology); anti-PolH (Santa CruzBiotechnology); anti-HA monoclonal antibodies (12CA5 andHA11) (Covance); and anti-Myc monoclonal antibody (9E10)(Cell Signaling).

Coimmunoprecipitation and RT–PCR

RKO cells that were transfected with a pcDNA3 vector express-ing HA-tagged RNPC1a, RNPC1a (F37S), or RNPC1b for 24 hwere collected and lysed at 4°C with a lysis buffer (50 mMTris-HCl at pH 7.4, 1% NP-40, 150 mM NaCl, 10 µg/mL apro-tinin, 10 µg/mL leupeptin, 1 mM PMSF, 2 mM Na3VO4, 50 mMNaF, 0.5 U/µl RNasin). Five percent of the cell extracts wereused directly for total RNA isolation, and the remaining ex-tracts were incubated with protein A/G beads conjugated withanti-HA or anti-Flag antibody (Sigma) overnight. Following fourwashes with a buffer containing RNase-free DNase, RNAs onthe beads were purified with a Trizol reagent. Reverse transcrip-tion was performed using iScript (Bio-Rad). To avoid potentialgenomic DNA contamination, both sense primer, 5�-ATGTCAGAACCGGCTGGGGATG-3�, and antisense primer, 5�-TTAGGGCTTCCTCTTGGAGAAGATCA-3�, were designedto amplify the entire p21 ORF (1 min at 94°C, 30 sec at 55°C,and 1 min at 72°C for 35 cycles). The primers used to amplifyGAPDH transcripts were forward primer, 5�-TGAAGGTCGGAGTCAACGGATTTGGT-3�, and reverse primer, 5�-CATGTGGGCCATGAGGTCCACCAC-3�.

Acknowledgments

We thank Anita Chen for technical assistance, Kelly Harms forthe MCF7 tet-on cell line, and Gang Liu, Jin Zhang, ArianeScoummane, Scott Helton, Yingquan Qian, and Yang Xu forsuggestions. This work is supported in part by NIH grants.

References

Barak, Y., Juven, T., Haffner, R., and Oren, M. 1993. mdm2expression is induced by wild type p53 activity. EMBO J. 12:461–468.

Battelli, C., Nikopoulos, G.N., Mitchell, J.G., and Verdi, J.M.2006. The RNA-binding protein Musashi-1 regulates neuraldevelopment through the translational repression ofp21WAF-1. Mol. Cell. Neurosci. 31: 85–96.

Brummelkamp, T.R., Bernards, R., and Agami, R. 2002. A sys-tem for stable expression of short interfering RNAs in mam-malian cells. Science 296: 550–553.

Shu et al.





Buckbinder, L., Talbott, R., Velasco-Miguel, S., Takenaka, I.,Faha, B., Seizinger, B.R., and Kley, N. 1995. Induction of thegrowth inhibitor IGF-binding protein 3 by p53. Nature 377:646–649.

Bunz, F., Dutriaux, A., Lengauer, C., Waldman, T., Zhou, S.,Brown, J.P., Sedivy, J.M., Kinzler, K.W., and Vogelstein, B.1998. Requirement for p53 and p21 to sustain G2 arrest afterDNA damage. Science 282: 1497–1501.

Chen, X., Bargonetti, J., and Prives, C. 1995. p53, through p21(WAF1/CIP1), induces cyclin D1 synthesis. Cancer Res. 55:4257–4263.

Chen, X., Liu, G., Zhu, J., Jiang, J., Nozell, S., and Willis, A.2003. Isolation and characterization of fourteen novel puta-tive and nine known target genes of the p53 family. CancerBiol. Ther. 2: 55–62.

Dohn, M., Zhang, S., and Chen, X. 2001. p63� and �Np63� caninduce cell cycle arrest and apoptosis and differentially regu-late p53 target genes. Oncogene 20: 3193–3205.

Dohn, M., Nozell, S., Willis, A., and Chen, X. 2003. Tumorsuppressor gene-inducible cell lines. Methods Mol. Biol. 223:221–235.

Dreyfuss, G., Matunis, M.J., Pinol-Roma, S., and Burd, C.G.1993. hnRNP proteins and the biogenesis of mRNA. Annu.Rev. Biochem. 62: 289–321.

Dreyfuss, G., Kim, V.N., and Kataoka, N. 2002. Messenger-RNA-binding proteins and the messages they carry. Nat.Rev. Mol. Cell Biol. 3: 195–205.

el-Deiry, W.S., Kern, S.E., Pietenpol, J.A., Kinzler, K.W., andVogelstein, B. 1992. Definition of a consensus binding sitefor p53. Nat. Genet. 1: 45–49.

el-Deiry, W.S., Tokino, T., Velculescu, V.E., Levy, D.B., Parsons,R., Trent, J.M., Lin, D., Mercer, W.E., Kinzler, K.W., andVogelstein, B. 1993. WAF1, a potential mediator of p53 tu-mor suppression. Cell 75: 817–825.

Giles, K.M., Daly, J.M., Beveridge, D.J., Thomson, A.M., Voon,D.C., Furneaux, H.M., Jazayeri, J.A., and Leedman, P.J. 2003.The 3�-untranslated region of p21WAF1 mRNA is a compos-ite cis-acting sequence bound by RNA-binding proteins frombreast cancer cells, including HuR and poly(C)-binding pro-tein. J. Biol. Chem. 278: 2937–2946.

Gorospe, M. 2003. HuR in the mammalian genotoxic response:Post-transcriptional multitasking. Cell Cycle 2: 412–414.

Gorospe, M. and Holbrook, N.J. 1996. Role of p21 in prostaglan-din A2-mediated cellular arrest and death. Cancer Res. 56:475–479.

Gorospe, M., Wang, X., and Holbrook, N.J. 1998. p53-dependentelevation of p21Waf1 expression by UV light is mediatedthrough mRNA stabilization and involves a vanadate-sensi-tive regulatory system. Mol. Cell. Biol. 18: 1400–1407.

Harms, K.L. and Chen, X. 2005. The C terminus of p53 familyproteins is a cell fate determinant. Mol. Cell. Biol. 25: 2014–2030.

Harms, K., Nozell, S., and Chen, X. 2004. The common anddistinct target genes of the p53 family transcription factors.Cell. Mol. Life Sci. 61: 822–842.

Harper, J.W., Adami, G.R., Wei, N., Keyomarsi, K., and Elledge,S.J. 1993. The p21 Cdk-interacting protein Cip1 is a potentinhibitor of G1 cyclin-dependent kinases. Cell 75: 805–816.

Johansen, F.E. and Prywes, R. 1994. Two pathways for serumregulation of the c-fos serum response element require spe-cific sequence elements and a minimal domain of serumresponse factor. Mol. Cell. Biol. 14: 5920–5928.

Joseph, B., Orlian, M., and Furneaux, H. 1998. p21(waf1) mRNAcontains a conserved element in its 3�-untranslated regionthat is bound by the Elav-like mRNA-stabilizing proteins. J.Biol. Chem. 273: 20511–20516.

Kaghad, M., Bonnet, H., Yang, A., Creancier, L., Biscan, J.C.,Valent, A., Minty, A., Chalon, P., Lelias, J.M., Dumont, X., etal. 1997. Monoallelically expressed gene related to p53 at1p36, a region frequently deleted in neuroblastoma and otherhuman cancers. Cell 90: 809–819.

Ko, L.J. and Prives, C. 1996. p53: Puzzle and paradigm. Genes &Dev. 10: 1054–1072.

Krecic, A.M. and Swanson, M.S. 1999. hnRNP complexes: Com-position, structure, and function. Curr. Opin. Cell Biol. 11:363–371.

Lal, A., Mazan-Mamczarz, K., Kawai, T., Yang, X., Martindale,J.L., and Gorospe, M. 2004. Concurrent versus individualbinding of HuR and AUF1 to common labile target mRNAs.EMBO J. 23: 3092–3102.

Lin, Y., Ma, W., and Benchimol, S. 2000. Pidd, a new death-domain-containing protein, is induced by p53 and promotesapoptosis. Nat. Genet. 26: 122–127.

Liu, G. and Chen, X. 2006. DNA polymerase eta, the product ofthe xeroderma pigmentosum variant gene and a target ofp53, modulates the DNA damage checkpoint and p53 acti-vation. Mol. Cell. Biol. 26: 1398–1413.

Liu, Y., Martindale, J.L., Gorospe, M., and Holbrook, N.J. 1996.Regulation of p21WAF1/CIP1 expression through mitogen-activated protein kinase signaling pathway. Cancer Res. 56:31–35.

Liu, G., Xia, T., and Chen, X. 2003. The activation domains, theproline-rich domain, and the C-terminal basic domain in p53are necessary for acetylation of histones on the proximal p21promoter and interaction with p300/CREB-binding protein.J. Biol. Chem. 278: 17557–17565.

Lopez de Silanes, I., Zhan, M., Lal, A., Yang, X., and Gorospe, M.2004. Identification of a target RNA motif for RNA-bindingprotein HuR. Proc. Natl. Acad. Sci. 101: 2987–2992.

Myer, V.E., Fan, X.C., and Steitz, J.A. 1997. Identification ofHuR as a protein implicated in AUUUA-mediated mRNAdecay. EMBO J. 16: 2130–2139.

Nozell, S., Wu, Y., McNaughton, K., Liu, G., Paik, J., and Chen,X. 2003. Characterization of p73 functional domains neces-sary for growth suppression. Oncogene 22: 4333-4347.

Shiohara, M., Akashi, M., Gombart, A.F., Yang, R., and Koeffler,H.P. 1996. Tumor necrosis factor �: Posttranscriptional sta-bilization of WAF1 mRNA in p53-deficient human leukemiccells. J. Cell. Physiol. 166: 568–576.

Siomi, H., Siomi, M.C., Nussbaum, R.L., and Dreyfuss, G. 1993.The protein product of the fragile X gene, FMR1, has char-acteristics of an RNA-binding protein. Cell 74: 291–298.

Timchenko, L.T., Miller, J.W., Timchenko, N.A., DeVore, D.R.,Datar, K.V., Lin, L., Roberts, R., Caskey, C.T., and Swanson,M.S. 1996. Identification of a (CUG)n triplet repeat RNA-binding protein and its expression in myotonic dystrophy.Nucleic Acids Res. 24: 4407–4414.

van de Wetering, M., Oving, I., Muncan, V., Pon Fong, M.T.,Brantjes, H., van Leenen, D., Holstege, F.C., Brummelkamp,T.R., Agami, R., and Clevers, H. 2003. Specific inhibition ofgene expression using a stably integrated, inducible small-interfering-RNA vector. EMBO Rep. 4: 609–615.

Vassilev, L.T., Vu, B.T., Graves, B., Carvajal, D., Podlaski, F.,Filipovic, Z., Kong, N., Kammlott, U., Lukacs, C., Klein, C.,et al. 2004. In vivo activation of the p53 pathway by small-molecule antagonists of MDM2. Science 303: 844–848.

Vogelstein, B., Lane, D., and Levine, A.J. 2000. Surfing the p53network. Nature 408: 307–310.

Vousden, K.H. and Lu, X. 2002. Live or let die: The cell’s re-sponse to p53. Nat. Rev. Cancer 2: 594–604.

Waldman, T., Lengauer, C., Kinzler, K.W., and Vogelstein, B.1996. Uncoupling of S phase and mitosis induced by anti-






cancer agents in cells lacking p21. Nature 381: 713–716.Wang, W., Furneaux, H., Cheng, H., Caldwell, M.C., Hutter, D.,

Liu, Y., Holbrook, N., and Gorospe, M. 2000. HuR regulatesp21 mRNA stabilization by UV light. Mol. Cell. Biol. 20:760–769.

Wu, X., Bayle, J.H., Olson, D., and Levine, A.J. 1993. The p53–mdm-2 autoregulatory feedback loop. Genes & Dev. 7: 1126–1132.

Wu, G.S., Burns, T.F., McDonald III, E.R., Jiang, W., Meng, R.,Krantz, I.D., Kao, G., Gan, D.D., Zhou, J.Y., Muschel, R., etal. 1997. KILLER/DR5 is a DNA damage-inducible p53-regu-lated death receptor gene. Nat. Genet. 17: 141–143.

Xiong, Y., Hannon, G.J., Zhang, H., Casso, D., Kobayashi, R.,and Beach, D. 1993. p21 is a universal inhibitor of cyclinkinases. Nature 366: 701–704.

Yan, W. and Chen, X. 2006. GPX2, a direct target of p63, inhibitsoxidative stress-induced apoptosis in a p53-dependent man-ner. J. Biol. Chem. 281: 7856–7862.

Yang, A., Kaghad, M., Wang, Y., Gillett, E., Fleming, M.D.,Dotsch, V., Andrews, N.C., Caput, D., and McKeon, F. 1998.p63, a p53 homolog at 3q27-29, encodes multiple productswith transactivating, death-inducing, and dominant-nega-tive activities. Mol. Cell 2: 305–316.

Yang, X., Wang, W., Fan, J., Lal, A., Yang, D., Cheng, H., andGorospe, M. 2004. Prostaglandin A2-mediated stabilizationof p21 mRNA through an ERK-dependent pathway requiringthe RNA-binding protein HuR. J. Biol. Chem. 279: 49298–49306.

Zhu, J., Jiang, J., Zhou, W., and Chen, X. 1998. The potentialtumor suppressor p73 differentially regulates cellular p53target genes. Cancer Res. 58: 5061–5065.

Zhu, J., Zhang, S., Jiang, J., and Chen, X. 2000. Definition of thep53 functional domains necessary for inducing apoptosis. J.Biol. Chem. 275: 39927–39934.

Shu et al.





10.1101/gad.1463306Access the most recent version at doi: originally published online October 18, 200620:2006, Genes Dev.

Limin Shu, Wensheng Yan and Xinbin Chen p21 transcriptrequired for maintaining the stability of the basal and stress-induced RNPC1, an RNA-binding protein and a target of the p53 family, is

Material

Supplemental

http://genesdev.cshlp.org/content/suppl/2006/10/17/gad.1463306.DC1

References

http://genesdev.cshlp.org/content/20/21/2961.full.html#ref-list-1

This article cites 50 articles, 24 of which can be accessed free at:

License

ServiceEmail Alerting

click here.right corner of the article or

Receive free email alerts when new articles cite this article - sign up in the box at the top

Copyright © 2006, Cold Spring Harbor Laboratory Press


http://genesdev.cshlp.org/lookup/doi/10.1101/gad.1463306

http://genesdev.cshlp.org/content/suppl/2006/10/17/gad.1463306.DC1

http://genesdev.cshlp.org/content/20/21/2961.full.html#ref-list-1

http://genesdev.cshlp.org/cgi/alerts/ctalert?alertType=citedby&addAlert=cited_by&saveAlert=no&cited_by_criteria_resid=protocols;10.1101/gad.1463306&return_type=article&return_url=http://genesdev.cshlp.org/content/10.1101/gad.1463306.full.pdf



Documents

RNPC1, an RNA-binding protein and a target of the p53 family, is …genesdev.cshlp.org › content › 20 › 21 › 2961.full.pdf · 2006-11-01 · bility of p21 transcript is found